Crude DNA extraction was performed by boiling a small loopful of fresh cultures for 10 min in a 2% (weight/volume) Chelex solution (Bio-Rad Laboratories Inc., CA) followed by centrifugation and decantation of the supernatant into a sterile tube with
storage at −20°C until PCR was performed. Each faecal isolate was initially tested by
C. upsaliensis/helveticus PCR (490) and each food isolate by C. jejuni (491) and C.
coli (492) PCRs; if negative followed by Campylobacter genus PCR (490).
Subsequent to a positive genus PCR, faecal and food isolates were sequentially
tested by species-specific PCR for C. lari, C. fetus, C. hyointestinalis (490), and
Arcobacter butzleri (493). A subset of isolates negative by the species-specific PCRs
was selected for 16S rRNA gene amplification and sequencing (494). Controls were used in all PCR assays and amplified products and 1Kb Plus ladder (Invitrogen Corp., USA) identified by electrophoresis in a 1% (weight/volume) agarose gel in Tris-Borate-EDTA buffer, subsequently stained with ethidium bromide and exposed to UV light using a Bio-Rad gel documentation system (Life Science Group, Canada).
The dog isolates confirmed as C. jejuni by PCR were genotyped using the multilocus
sequence typing scheme (MLST) of seven housekeeping genes (256). The
amplifications were performed in a 25μL volume reaction using Applied Biosystems
AmpliTaq Gold mastermix (Applied Biosystems, Auckland New Zealand) and 5 pmoles of each primer. Products were sequenced on an ABI 3130XL automated DNA sequencer using ABI BigDye v3.1 (Applied Biosystems) following the manufacturer's instructions. Sequence data were collated and alleles and sequence
types (ST) assigned using the CampylobacterPubMLST database (URL
http://pubmlst.org/campylobacter/) and occurrence compared with other sources in
mEpiLab database. The mEpiLab database contains over 3,500 samples (at the time
of writing) from the Manawatu Campylobacter sentinel site, a ten plus year project for
source attribution of campylobacteriosis using concurrent sampling of human cases, animals, food and the environment (312).
124 4.4.4. Statistical analysis
Results of culture methods were compared using Fisher’s exact test of independence. Statistical and exploratory data analyses were performed using R v3.2.2 (R: A language and environment for statistical computing. R Core Team
(2013). R Foundation for Statistical Computing, Vienna, Austria. URL http://www.R-
project.org/).
4.5. Results
From a total of 50 dog faecal samples, there were 408 presumptive Campylobacter
isolates with 356 testing positive by Campylobacter genus PCR, of which species-
specific PCRs returned 232 C. upsaliensis, 81 C. jejuni, 14 C. coli, and one C. lari.
Five isolates were PCR positive for A. butzleri. Fifteen isolates with negative species-
specific PCRs from various dogs and from every culture method, as available,
returned five Helicobacter winghamensis, four C. upsaliensis, two each of C. rectus
and C. volucris, one C. lari subsp. concheus and one A. cryaerophilus as the most
similar species by 16S rRNA sequencing. The remaining eight isolates could not be
identified. A relative comparison in performance between culture methods is
presented in Table 4.1. Overall, two dogs (4%) were positive for three Campylobacter
spp., five (10%) for two and 24 dogs (48%) for one species only. Three dogs (6%)
were positive for Campylobacter spp. by one method only, six (12%) by two, nine
(18%) by three, ten (20%) by four, and three (6%) dogs by five methods. Combining
all the methods, 24 (48%) dogs were Campylobacter spp. positive on day two, four
(8%) on day three, two (4%) on day four and one (2%) on day five (C. lari) of
incubation.
125
Table 4.1. Number of positive working farm dogs1 from Manawatu, New Zealand (N = 50) using seven culture methods and isolates identified by PCR.
Method2 Campylobacter spp. C. upsaliensis C. jejuni C. coli Other
CB_H2_AB 2 a - a - a 1 1 CB_H2_CAT 8 b 7 b, c - a - 2 CAT_MA 25 c 21 d 5 a, b - 2 CAT_ H2 24 c 18 d 5 a, b 1 2 BB_ H2_mCCDA 21 c 8 b 8 b 3 5 BB_MA_mCCDA 6 a, b 3 a, b 2 a, b 1 - AB_H2 11 c 8 c, d 3 b - - Overall 31 21 9 3 8
1Shared superscript letters within each column denote no significance by Fisher’s
exact test (α < 0.05). 2CB (non-selective enrichment broth); BB (selective Bolton
broth); AB (non-selective anaerobe basal agar); CAT and mCCDA denote respective
agars; MA (microaerobic); H2 (H2-enriched MA). All methods performed at 37°C
except BB_MA_mCCDA at 42°C. CB_H2_CAT used on 38 and AB_H2 on 21 dogs.
Species isolated for the first time in dogs were as follows: C. volucris by the methods
CB_H2_CAT and CAT_MA (on fourth and second day of incubation respectively), C.
lari subsp. concheus by the method BB_ H2_mCCDA (on fourth day of incubation),
C. rectus by the method CB_H2_AB (on fourth day of incubation) and H. winghamensis by the methods CB_H2_CAT, BB_ H2_mCCDA, and CAT_ H2 (on
126
(13/50), 79% (30/38), 98% (49/50), 96% (48/50), 100% (50/50), 98% (49/50), and 100% (21/21) with the methods CB_H2_AB, CB_H2_CAT, CAT_MA, BB_H2_mCCDA, BB_MA_mCCDA, CAT_ H2, and AB_H2 respectively, and selectivity was 55% (6/11), 80% (16/20), 93% (97/104), 94% (72/77), 100% (20/20),
93% (104/112), and 63% (41/65) respectively. Forty C. jejuni isolates from eight dogs
(two to nine isolates per dog) were subjected to MLST typing and 33 returned full allelic profiles. Eight different STs belonging to five different clonal complexes were observed and two dogs carried STs of different clonal complexes. The occurrence of
these ST in other sources from the mEpiLab database is depicted in Supplemental
Fig. 4.1.
From 50 home-kill meat samples, there were 52 presumptive Campylobacter isolates
with 17 testing positive by Campylobacter genus PCR from three samples (6%). Of
these 17, four isolates were positive by C. jejuni PCR and all were grown using
mBB_MA_mCCDA from one meat sample (2%) while one other sample (2%) grew 11 isolates using mCB_H2_CAT, mCB_MA_CAT, mBB_H2_mCCDA, and
mBB_MA_CAT, which were all positive for A. butzleri by PCR. Sequencing of the
16S rRNA gene for the two remaining isolates returned C. rectus as the most similar
species isolated using mCB_MA_CAT in the third meat sample (2%). The proportion of readable plates was 38% (19/50), 89% (34/38), 94% (47/50), 100% (50/50), 100% (50/50), and 98% (49/50) with the methods mCB_H2_AB, mCB_H2_CAT, mCB_MA_CAT, mBB_H2_mCCDA, mBB_MA_mCCDA, and mBB_MA_CAT
respectively. The presence of E. coli was detected in 48% (24/50) of the meat
samples.
4.6. Discussion
The main findings of the study are the significant differences in isolation of
Campylobacter spp. between the culture methods and the isolation of four species
previously either not reported or not isolated from dogs. These results are of potential public health importance as all species identified in dogs and their food are
127
described in black-headed gulls (495) and since then reported in an
immunocompromised human patient with bacteraemia (496). C. lari subsp. concheus
was initially isolated from shellfish and subsequently in humans, seagulls and river
water (66, 497). C. rectus was detected previously using molecular methods in dog
faeces (359) and oral swabs (487) but the clinical significance is uncertain in dogs. In
people, C. rectus is associated with periodontitis/gingivitis, various gastrointestinal
diseases and extra-intestinal infections and, apart from dogs, no other potential
sources have been identified (13). H. winghamensis was described as a novel
species upon isolation from people with clinical signs of gastroenteritis (498) and so far only one study using molecular methods has reported its occurrence in animals and rodents (499). All of the above species have been rarely reported, thus it is not clear if the animals are the true reservoirs for the organism or if they are just transient
carriers. With regard to the faecal carriage of C. jejuni and C. upsaliensis, a
longitudinal study in dogs reported carriage of C. jejuni of short duration and with
genotypically diverse isolates using pulse-field gel electrophoresis whereas the
carriage of C. upsaliensis was of long duration of clonal strains (139). Four out of
eight C. jejuni STs isolated in this study are very rarely observed in the mEpiLab
database and the other four STs are common in several sources which supports the heterogeneous exposure of farm dogs (Supplemental Fig. 4.1). Other studies using MLST also reported a high diversity of STs in dogs including strains frequently seen
in humans and food (300, 477). In contrast to this study, Campylobacter spp. were
isolated from 13% and C. jejuni from only 5% of 498 dog faecal samples in the
Palmerston North area (mostly dog walking areas) (356). In that study, a culture method similar to the BB_MA_mCCDA used in the current study. Lower prevalences using the BB_MA_mCCDA method alone compared to other methods and overall results in this study (Table 4.1) suggest the different results between the two studies are due to the culture methods used. However, the two studies also had different dog populations, faecal sample handling and sample sizes that make results less directly comparable.
The benefit of applying multiple culture protocols in this study is evident from the significant differences observed in pairwise comparison of methods, both in the
128
the less common species A. cryaerophilus, C. lari subsp. concheus, H.
winghamensis and C. rectus, the common denominator was isolation in H2-MA, while
for C. volucris the use of CAT agar and for Arcobacter spp. use of BB_H2_mCCDA
appeared to be the most suitable. The requirement of hydrogen for isolation of many
emerging Campylobacter spp. as well as the enhanced recovery of C. jejuni has
been previously recognised (17, 18). Interestingly, C. rectus isolated from meat
samples in this study grew in pure MA although it is considered to have a requirement for hydrogen to grow (36). In this study, too few of the emerging species
were isolated for statistical comparison, but with regard to C. jejuni and C.
upsaliensis, CAT_MA and CAT_H2 only differed in the presence of hydrogen and no
significant differences were observed (Table 4.1). Although not statistically significant
(p = 0.06), a surprising finding was the difference in isolation rate of C. jejuni between
BB_H2_mCCDA and BB_MA_mCCDA (Table 4.1). The two methods differ by
temperature and presence of hydrogen, and the enhanced isolation of C. jejuni at
37°C rather than at 42°C with mCCDA has been reported (96). C. upsaliensis had
the largest difference in isolation rate between methods in this study (Table 4.1).
Comparison of CAT and mCCDA for isolation of C. upsaliensis in veterinary studies
are conflicting as a higher isolation rate with CAT compared to mCCDA has been reported (54, 94) but also equivalent rates (402), and a lower rate with CAT (405) for
both C. upsaliensis, and C. jejuni. Similarly to the current study, in human studies
mCCDA was outperformed by both CAT (104) and by the filtration method (96) for
the isolation of C. upsaliensis. In contrast, filtration was reported as superior to CAT
(95) but in the present study the two were comparable (Table 4.1). CAT was
originally developed according to the antimicrobial resistance profiles of several
thermophilic Campylobacter spp. (104) but was also shown to result in better growth
and isolation of a greater diversity of C. upsaliensis strains than mCCDA (472) and
enhanced detection of lower bacterial concentrations compared to mCCDA (99). However, the latter study could not explain the difference in sensitivity between the two agars, neither by the absolute growth index for any length of incubation time nor the antimicrobial composition of the media, leading investigators to speculate that the
growth of C. upsaliensis is indirectly affected by the interaction of faecal microflora
129
a comparable rate of isolation of C. upsaliensis in CB_H2_CAT with both BB/mCCDA
methods was observed, while direct plating on CAT outperformed all three of them
(Table 4.1). Improved recovery and motility of C. upsaliensis in CB compared to BB
has been reported (43). These observations could explain the poorer performance of methods using BB in this study but not that of CB. The poor performance of CB_H2_AB is likely to be due to overgrowth of contaminants, as the modification to CB_H2_CAT both improved the readability of plates and the isolation of
Campylobacter although the isolation success was still significantly less than that of
other methods (Table 4.1). This suggests that contaminating organisms may inhibit
Campylobacter cells in CB. In this study only one dog was identified as positive after
four days of incubation but the relatively low number of samples is likely to preclude weighting of this observation and incubation up to six days has been advised for higher isolation success (18).
The main objective for this study was to isolate a diverse set of Campylobacter spp.
using the novel culture method, CB_H2_AB, which, with the exception of C. rectus
being isolated from one dog, failed in both faecal and meat samples. The failure was
primarily ascribable to frequent overgrowth by contaminants (mostly Proteus spp.
and less frequently Pseudomonas species). The CB_H2_AB was developed on fresh
beef samples only, with no report of overgrowth by contaminants (43). Perhaps the home-kill meat (48% of which showed faecal contamination) and the dogs’ faeces used in the current study contained too many non-target bacteria, making the antimicrobial-free method unsuitable. Overgrowth by contaminants on plates using the filtration method (without enrichment) and using selective plates was reported, but usually in less than 10% of plates (94, 99) which is similar to methods using antimicrobial agents (faeces and meat) used in this study. CB_H2_AB with an enrichment duration reduced to 24 hours was also successfully applied previously in many types of fresh meat products (61) and porcine samples including caecal contents (62). As the overgrowth of contaminants in this study could not be explained by the procedures in production, storage and usage, the modified (m)CB_H2_CAT methods were added during the study. The readability of the plates rose to 79% and 89% with faeces and meat respectively. In addition, increasing the
130
in 12 samples, but was unsuccessful. To investigate the overgrowth of contaminants associated with the CB enrichment rather than the filtration technique in CB_H2_AB vs. swab to CAT in CB_H2_CAT, the AB_H2 method was added during the study. Although AB_H2 was only applied to 21 faecal samples the readability rose to 100% which suggests CB caused overgrowth of contaminants to a level too high for the filtration method and partially too high for the CAT to cope with.
Another modification that could be useful in optimising CB_H2_AB and other
methods in this study used to isolate Campylobacter from food samples is the
inclusion of a pre-enrichment step (usually up to 4 hours) with either delayed addition of antimicrobials to broth, a lower incubating temperature or both. This step is
suggested when low number or injured Campylobacter cells are expected, such as in
frozen products (20, 110). The absence of pre-enrichment step in the current study
could explain the low prevalence of Campylobacter spp. observed in frozen home-kill
meat by all methods used. There were 44 sheep and 6 beef samples in this study
and the prevalence of C. jejuni in retail fresh meat or liver for human consumption in
New Zealand is 25% and 8% in sheep and beef meat respectively (312). It could be expected that home-kill meat has a greater contamination level than commercially available meat due to a lack of hygienic measures in farming environments and considering the intended use of the two meat types with the regard to preparation, handling, and storage practices. However, the sample size of this study is low and limits the confidence for comparison of the studies.
The selectivity estimates were used to describe the workload with the different culture methods given the protocol for identification of species used. The rationale
was, that suspect Campylobacter colonies that were negative by Campylobacter
genus PCR were subcultured, stored and re-tested for no benefit for the time and
resources invested. The variation in colony morphology of Campylobacter spp.
between agar plates was reported (38) but is not related to selectivity expressed herein which denotes the ratio of PCR-confirmed over presumptive colonies. Low
selectivity suggests the isolates should not be considered Campylobacter spp. and
should be confirmed by further identification tests. Lower estimates were observed with filtration methods and CB than with the use of antimicrobial agents. It is possible
131
the antimicrobials have supressed a wider range of species than the filtration method, thus providing a lesser diversity of bacteria on the agar, of which even fewer
were Campylobacter-like. In contrast, a greater diversity of bacteria passed through
the filters, many of which grew on a non-selective agar and more of which were
Campylobacter-like. However, the aim of the study was to isolate a variety of Campylobacter spp. hence screening of isolates that otherwise may not be included
was expected. C. showae resembles straight rods (44), C. mucosalis has yellow
coloured colonies (36), and C. gracilis is non-motile (45); all are examples of isolates
that would not be included if the focus was strictly on the phenotypic characteristics of the common species. Additional biochemical or phenotypic tests (47) could have been applied for presumptive isolates in this study, which could change the selectivity estimates by reducing the number of isolates passed to PCR testing. However, the addition of more screening tests for isolates increases the workload and cost. The
cross-reaction of the PCR for Campylobacter spp. with Arcobacter spp. has been
reported (474) but the cross-reaction with H. winghamensis is newly observed.
These are closely related genera within Campylobacteraceae that can be isolated
using similar culture methods (92). The taxa have relatively high similarity both phenotypically and genotypically (500) which makes their cross-reaction less surprising.
Conflict of interest statement
The study was funded by the Centre for Working and Service Dogs, Massey University, New Zealand and authors declare no conflict of interests.
Acknowledgments
The authors would like to thank their farmers for participating in the study, Cristobal Verdugo and Anou Dreyfus for use of the farm database and Prof Boyd R Jones for providing transport to the farms and for critical reading of the manuscript. The
isolates in the mEpilab database were collected as part of the campylobacteriosis
132